The traditional B2B marketing funnel isn’t dead, but the buzzards are taking bets. Really, the top is stretching (and drifting away from you, the brand owner), the middle is shrinking, and way down at the bottom, buyer intent is arriving late to the party, fully briefed and ready to go.
So…how do you influence buyer behavior now?
In this post, I’ll theorize a revised funnel, explore some industry-specific nuances, and break down how all of this is going to impact marketing attribution and reporting. (Spoiler alert: you better get comfy with gray areas.)
Contents:
Why Is the B2B Marketing Funnel Breaking?
How The Changing Funnel Hits SaaS, Pro-Serv & Manufacturing
How to Handle Measurement & Attribution in the New B2B Funnel
In 2026, the Buyer Is In Control & Signals Take the Lead
Frequently Asked Questions on the Modern B2B Marketing Funnel
Inbound marketing has been the ten-ton gorilla of marketing models for the last ten-plus years. And why not? It has a neatly packaged funnel and a clear driver of leads (hello, gated content). But let’s face it: the lead math hasn’t been mathing for a while.
Even the company who pioneered inbound marketing (and has an annual event named after it) is now rethinking the funnel approach with a totally new shape (the “loop”).
At Conveyor, we don’t believe the funnel is dead. But it is showing stress fractures.
Content still has a TON of value, but it isn’t about using a form to “qualify” leads. Making every form fill a marketing qualified lead (MQL) led to too many “leads” getting routed to sales long before they’re ready.
The buyer’s journey is evolving: research is happening via AI-assisted search, on community-driven channels and forums, and by way of trusted trade pubs or user reviews. You can’t (easily) see or measure a lot of this activity, because it’s not happening on your site.
Traditional search is changing by the actual minute. Search engine optimization (SEO) is not dying; I would argue it’s more important than ever in a world of AI-assisted search (as is brand awareness), and generative engine optimization (GEO) is not a totally new animal, no matter what this week’s LinkedIn influencer tells you. However, informational search traffic is collapsing due to AI Overviews and zero-click search. These queries still generate visibility, but far fewer clicks, meaning any top-of-funnel strategy must focus on being cited and visible in AI summaries, while reallocating media budget to transactional keywords and off-site influence.
The traditional models were always limited, but with ever-increasing privacy restrictions and much of the buyer research happening off-site, the mid-funnel is shrinking. It’s harder than ever to connect the top of your funnel to the bottom. Dashboard-perfect attribution is now a myth (if it wasn’t always).
Bottom line: Traffic growth and form fills are no longer the measure of success. Your models need to change.
Like I said, I don’t think the funnel is dead. But it needs a facelift. Never fear: there is a new framework.
The top of the funnel is expanding (quickly) on channels and platforms you don’t own. Like it or not, we have to face that. Community-driven conversation at the top is critical to building your brand, but when you engage here, it has to be authentic and done by real people from your org, not a “corporate profile.”
Common channels at this level include Reddit, G2, LinkedIn, Substack, YouTube, trade pubs, AI/LLM chats, Google search results, and much, much more.
This is the part of the funnel that’s stretching like the entry parlor in The Haunted Mansion. To help get your head around it, we recommend breaking ToFu into more granular mental “chunks”:
Technically, this is outside of the funnel. The companies and people here may never be your customer, but they could be a critically important voice (cough, cough, influencer or key opinion leader) in your space that could make or break you. Even though you can’t tie this group to revenue, don’t underestimate the importance.
Your ICP (ideal customer profile) and TAL (target account list) are your best prospects. They are the people you need to notice your brand, and you are willing to pay for their attention. You’re targeting them with advertising, outbound prospecting, and more, because they are your perfect customer. But…they’re not engaging with you. Yet.
Your audience comes from within your ICP + TAL; these are the people who are actively engaged with your brand, but are not yet in your owned ecosystem. They may be unknown website visitors, social media engagers, or ungated/off-site content consumers. They know you, you might know them, but you’re not dating. And they haven’t yet taken an action that shows intent.
This group knows you and you know them – and they know you know them: they’ve signed up for a newsletter (possibly on your site, but just as likely via LinkedIn or Substack or similar), attended a webinar (possibly one you hosted, but maybe one with a media partner), or downloaded a gated thought leadership piece such as first-party research. They may also be engaging with your outbound prospecting, but they are NOT YET showing intent. However, you can now nurture them.
AI Overviews are decimating informational traffic, the fuel that once fed the top of the funnel and flowed leads into mid-funnel content, and that’s a big part of why the mid-funnel as we once knew it is shrinking. Given how much education is happening off-site today, a site visit has become a pretty intentional decision. We’re now considering those visitors your mid-funnel.
The role of the marketing qualified lead (MQL) is also changing as the mid-funnel shrinks, and it may one day stop being a useful stage altogether. The jury’s out for the future, but for now, the mid-funnel still serves a clear purpose: identifying buyer intent.
This used to be any lead that took a “hand-raising” action like filling in a form or someone that hit a target score based on content engagement. Today, MQL should reflect a key action that shows intent: accepting an offer such as requesting a demo or a consultation, or looking at some combination of key content (ex: reading your pricing page, looking at your “how we work” page, and downloading a product sheet all in one session). MQLs should be flagged to sales immediately with a service level agreement (SLA) to take action in 24–48 hours.
These MQLs have been manually vetted by a human in sales. This stage is intentionally short. It may alternatively be named sales accepted leads (SALs).
Less has changed here, at least so far. Once a lead has shown intent and a salesperson has agreed that the lead is real, they're at the bottom of your funnel, and while they’re here, they’ll fall into one of these categories:
While the funnel is changing across all B2B (business-to-business) sectors, I’m going to take a deeper dive into just a few of the industries most impacted here.
SaaS marketing has always been a volume play. Quality mattered, but more so further down the funnel; the ToFu focus was generating enough traffic so you could drive more of the mid-funnel conversions needed to support sales.
But non-branded informational search queries are drying up, and the easy exposure you used to get from your ToFu content is harder to come by. Now, off-site signals (user reviews, Reddit discussions, social media chatter) look like SaaS’s best path forward.
We’re focused on intent: weighing your offsite signals alongside TAL/ICP fit, and using your website to qualify visitors based on behavior. The focus of measurement should shift to an account-level scorecard that tracks the activity of entire buying committees across key stages. (You can find an example near the end of this article.)
Traffic and lead volumes typically matter less in pro-serv marketing; quality is the gold standard. And unlike SaaS, pro-serv is less exposed to the AI Overview informational search collapse, because referrals tend to be the lifeblood of sales here. But as referral networks lose momentum, the best play is ToFu, with executive thought leadership as a key strategy.
That’s because pro-serv companies are really selling the brainpower of their team, and that requires building influence, communicating credibility, and positioning around expertise. Featuring their experts as thought leaders gives pro-serv companies access to new and growing audiences. But attribution challenges are just as daunting.
We suggest leaning into authority signals: things like speaking invitations, citations, and media mentions. These are leading indicators of credibility and recognition, and ultimately become proof points for your brand.
In this industry, the website has traditionally been a brochure. It’s now a BOF validation check once someone is fairly far along in the buyer’s journey. Use your site to capture their intent.
In manufacturing marketing, inbound once dictated that operational proof points should be hidden behind forms for the sake of generating “leads.” But that was never the best strategy; those leads took ages to nurture (and many never became customers). Now that the role of your website has shifted, hiding your operational proof is not an option; you need them easy to find if you want to educate prospects in this sector.
In 2026, kill all of your mid-funnel forms and only keep valuable first-party research gated. All other content should be open; forms are now for BoFu intent offers only.
Then, bring your proof forward. Pull data from your enterprise resource planning (ERP) system (look at things like downtime reductions, throughput increases, scrap reductions, etc.) and surface them on your website with live proof blocks that refresh on a regular cadence. Aggregate and anonymize as needed. If your data isn’t mature yet, publish process proofs (windows + variances) instead.
Remember: your website is now about qualifying by intent signals, so offer no more than three BoFu assets max. Create calculators, checklists, or price teardowns, then reach out as soon as someone converts.
How to Handle Measurement & Attribution in the New B2B Funnel
So the funnel has changed. How does that impact the math, the modeling, the budget?
The Old Models Were Never Great; Now They’re Worse
For years, marketing attribution relied mostly on first-touch, last-touch or linear models (among others). Why? Because even when we use tools that can deliver a more sophisticated model, they are still difficult to make useful decisions from.
And these models worked ok when traffic and form fills equaled success.
Now, they’re too simplistic:
They more or less sprinkle attribution arbitrarily (frustrating—and a tongue twister).
Everything is upside down now.
Even in high-volume plays like SaaS, traffic volume matters less than who is visiting your site (TAL or ICP fit); the so-called vanity metrics of yesteryear like brand mentions, social following, user reviews, newsletter subscriptions, and off-site engagement are now better leading indicators.
And yet, digital marketing and automation got us all addicted to 1:1 attribution, tracking every click across the funnel, all the way to revenue. That mindset overcredits visible channels like ad clicks while ignoring the harder-to-track stuff, like community engagement and brand recognition.
We’re always under pressure to tie dollars out to dollars in. But that’s how we ended up with broken attribution models in the first place.
Now, it’s getting worse:
In many cases, it’s not that your funnel’s broken. It’s that you’re paying to prop up ToFu informational queries Google no longer values.
Bottom line: “Exact attribution” is a lie. The world changed—your model needs to, too.
When using an ABM approach, account-level scorecards seem to be a better potential model for measurement in a world where progress matters more than perfection. We’re testing a model similar to the one below as a way to track engagement by account and buying committee.
|
Exposure 1 point |
Engagement 2 points |
BoFu Conversion 3 points |
|
|
|
|
Boost for Fit Boost points for role fit and/or buying committee reach. For example, if your exposure is with a decision maker, maybe give 1.5 points. Or, if you’re proving engagement with an entire committee, boost by 2X. Cap your boosts at +2 points. |
||
If you’re targeting by ICP fit instead of (or in addition to) an ABM play, then the scorecard above is a bit too prescriptive. The same overall principles apply in terms of engagement stages, and you can use your lead scoring models to capture that.
What really needs to change here is how we think about the ToFu in general. As we continue to lose visibility into where and how leads first find out about your brand, you’ll need to lean into monitoring trends as a proxy.
If you’re investing in brand-focused campaigns or thought leadership strategies (really, any brand-building activity) and you’re seeing an increase in lead quality, then you should continue your ToFu efforts even if you can’t clearly tie them back to revenue.
Monitor trends over time and look for correlations in spikes and timing. For example, a branded campaign may not deliver much in terms of clicks, but perhaps you see an increase in direct traffic that corresponds to the timeframe of the ad campaign. Or maybe every time your CEO is interviewed by a particular podcast, you see a lift in social followers or newsletter subscribers.
Yes, five years ago, we would not have recommended this. But again, the world has shifted, and the things that we once considered vanity metrics are becoming more important signals.
So how can you sell this to ROI-obsessed leadership? This is where your financial metric guardrails come in. Keep a close eye on CPL and the cost of a new customer. If those are moving in the wrong direction, you know you need to make changes. But if they’re improving and your trend lines are up – then you’re doing something right!
First, let’s stop obsessing over MQL math. As the funnel stretches and the emphasis moves to the top and bottom, MQLs may weaken as an indicator of funnel strength.
Shift your focus from form fills to intent signals. These signals (and the most appropriate next actions they demand) should be based on individual visitor behavior.
It’s time to build content ecosystems that do more than hide PDFs behind forms. Looking ahead, your content must:
The days of pushing buyers through a company-defined nurture process are over. The buyer is in control. The best we can do is meet them where they already are, listen closely to what they’re doing, and be ready to move when they are.
Looking for a comprehensive funnel review? Schedule a working session with me and my team.
No, the B2B funnel isn’t dead, but transformed. The top has stretched (buyers research off-site), the middle is starting to evaporate, and intent shows up later, and much closer to the bottom, as leads arrive at your site pre-educated.
Because a form fill doesn’t equal buying intent anymore, MQLs don’t mean as much. Real signals, like demo requests, ROI calculator use, or pricing page visits, matter more.
Ungated proof points, thought leadership, and value-driven resources are the modern replacements for gated content in the new funnel; they educate buyers while letting them self-qualify.
Track account-level signals across exposure, engagement, and conversion, and tie results to financial metrics, win rate, and meetings booked to start measuring success in the new funnel.
Your website has shifted to the mid-funnel. It acts as a qualification hub to uncover intent, not just a traffic collector.
AI supported the development of this content, including planning, brainstorming, and outlining, but a human did the writing (and editing).